STAT5 Has Tumor Suppressor-Like Activity in a Murine Model of Myc-Initiated Acute B-Lymphoblastic Leukemia/Lymphoma

Abstract 919

Signal transducer and activator of transcription 5 (STAT5) is a critical regulator of normal and leukemic lympho-myeloid hematopoiesis through activation downstream of early-acting cytokines, their receptors, and janus kinases (JAKs). Despite upstream activating mutations driving JAK-STAT phosphorylation in precursor-B acute lymphoblastic leukemia (B-ALL), activated JAK-STAT is absent from the aggressive “double hit” lymphomas expressing myc and bcl-2. Using C57BL/6 background transgenic mouse models for myc and bcl-2, we set out to determine whether endogenous STAT5 functions in guarding against B-ALL induced by combined myc/bcl-2 or myc alone. We first determined whether constitutive expression of bcl-2 driven from the H2K promoter and Moloney murine leukemia virus enhancer in C57BL/6 background STAT5-deficient hematopoietic cells could bypass blocks in B-lymphocyte development. Transgenic H2K/bcl-2 expression in hypomorphic STAT5ab^DN/DN mice, which are leaky and still produce some mature B-lymphocytes, largely rescued peripheral B-lymphocyte survival to near normal levels but could only rescue about 10% of the multilineage hematopoietic stem cell (HSC) competitive repopulating defect. Complete deletion of the entire STAT5ab locus resulted in the expected severe block of B-cell development at the pre-pro-B-cell stage following transplantation of STAT5ab null/null fetal liver cells into irradiated wild type or common γC^−/− recipients. Peripheral B-lymphocyte development could not be restored by transgenic bcl-2 alone in the absence of STAT5. However, transgenic myc driven from an immunoglobulin H chain enhancer (Emu/myc) combined with H2K/bcl-2 induced B-ALL peripheral counts as high as 1.1 × 10⁵ B-cells/ul and reduced latency (a median survival of 44 days) compared to wild-type control (a median survival of 91 days) in either lethally-irradiated (P<0.001; N range from 8–14 mice/group) or sub-lethally-irradiated cohorts of fetal liver transplanted mice (P=0.007; N range 10–20 mice/group). B-ALL in mice with or without STAT5 was a mix of Pro-B and Pre-B ALL (IgM-CD43+B220+CD19+/−CD4+/−) and morphologically similar in the spleen and bone marrow. Multi-parameter flow cytometry analysis of bone marrow cells from STAT5ab null/null fetal liver transplanted mice (N=4) showed that deletion of STAT5 significantly reduced by 11.5-fold (P=0.004) the fraction of long-term repopulating HSC (CD150+CD48-) c-Kit+Lin-Sca-1+ (KLS). In an independent adult Mx1-Cre conditional knockout of STAT5 by pI:pC treatment, lymphomas induced by Myc alone were also accelerated (P=0.05; N range 14–15 mice/group) with STAT5 maintained deleted in sorted B-cells. These mice also had reduced CD150+CD48- KLS cells (5.6-fold; N=4; P=0.006). Interestingly, several phenotypes recently reported as associated with increased HSC cycling and lymphoid-biased differentiation were observed. The mean fluorescence intensity of slamf1 (CD150) was reduced 2.2-fold (P<0.001; N=4) in conditional knockout mice and the B-lymphoid biased CD48+CD150+ or CD48-CD150- KLS cells representing short-term HSC/multipotent progenitors were not significantly reduced. Microarray analyses of the KLS fraction provided evidence that STAT5 promotes HSC maintenance and myeloid potential (limiting lymphoid commitment, cycling) in the KLS compartment. The deletion of STAT5 reduced expression of HSC self-renewal and quiescence promoting genes and increased immunoglobulin and B-lymphoid transcripts. Combined with the pre-pro-B-cell block, loss of STAT5 promotes accumulation of B-lineage committed progenitors as potential ALL initiating cells. The effects of bcl-2 and myc hits on STAT5 null/null hematopoietic cells are currently being further characterized with respect to B-cell developmental blocks and molecular heterogeneity. B-ALL has a high relapse rate and is driven by clonally diverse tumor propagating populations. Our work may have important implications for ALL drug therapy. In conclusion, we demonstrate that STAT5, considered primarily as functioning like an oncogene in hematologic malignancies upon persistent activation, can play a tumor suppressor-like role in subsets of B-ALL. These data add to an emerging understanding that endogenous STAT5 can suppress some cancers and transcriptionally regulate several cell cycle inhibitors.